Osborn. — Some Observations on the Tuber of Phy 1 1 oglossu m . 499 
These subsequently proved to be the growing-point in process of invagina- 
tion. 1 The third leaf showed signs of damping off, but it had a short stalk 
bearing a slightly swollen tuber, enclosed in which the growing-point could 
be discerned. No rhizoids were present at this stage. All the others 
appeared healthy, but showed, with one exception, little development above 
ground. The exception had two ‘ leaflets 1 about 4 mm. long, one arising 
on each side of the leaf and springing from the lower side. Unfortunately, 
when examined three or four days later, this leaf had collapsed too much to 
draw. Each leaflet appeared to be attached directly to the main leaf, with 
no noticeable swelling on or beneath the surface of the soil. 
Text-figs. 23-29. Leaves of various series. The dates below in brackets are those on which 
the various figures were drawn. Allx5^. Fig. 23. Leaf of Ser. C. The cell mass is nodulose; 
a T.S. of this leaf is seen Fig. 38 (Sept. 24). Fig. 24 a and b. Leaf of Ser. C. This leaf curved 
on soil in horizontal plane. Adventitious growth from convex side (Nov. 16). Fig. 25. Leaf ot 
Ser. D. Note absence of definite growing-point, and formation of many starchy cell masses 
(Oct. 21). Figs. 26-29. Leaves of Ser. E; for further explanation see text (Nov. 25). 
By November 16, almost four months after the experiment was started, 
the remaining leaves were seen to be yellowing rapidly. They seemed to 
be mere hollow shells for the greater part of their length, and collapsed 
when lifted. The base of the leaf, however, was still green, as was the 
small cell mass. The leaves were all curved, four with their apices erect ; 
the fifth lay curved C-shaped on the soil. In this case also the adventitious 
growth was from the convex side and so placed that the cell mass originally 
would not have been in direct contact with the soil (Text-fig. 34 A and B). 
Each of the leaves had developed a short stalk bearing a tuber, on which 
rhizoids were just appearing. 
See p. 504 below. 
